2. A model fit predicting preference for U.S. President (Democrat, Republican, In- dependent). using r = annual income (in $10,000) is log(i,/i,) = 33 -0.2r and log(i,/i,) = 1.0 + 03x. Plot ap, i,, and ap for I between 0 and 10, and interpret. (Hint: The logits and an ta, tag = 1). [15 pts] 3. The following table refers to a clinical trial for the treatment of small-cell lung cancer. Patients were randomly assigned to two treatment groups. The sequent tial therapy administered the same combination of chemotherapeutic agents in each treatment cycle; the alternating therapy had three different combinations, alternating from cycle to cycle. Response to Chemotherapy Progressive No Partial Complete Therapy Gender Disease Change Remission Remission Sequential Male 28 45 29 26 Female 4 12 2 Alternating Male 41 44 20 20 Female 7 (a) Fit a cumulative logit model with main effects for treatment and gender. Interpret. (If you use functions Irm or polr, the weights option may be used, depending on how you organize the data. If the data is organized in the same way as the mental impairment example in the lecture, the weights option is not needed. If the counts of each combination of the three variables are contained in the data, the weights option should be the counts). [15 pts] (b) What are P(Y